In order to be able to rationally design oligonucleotides which are stable in culture and penetrate cells more efficiently than normal oligos, we have investigated the mechanism by which cells transport normal oligomers. Using acridine-labeled oligos and flow cytometry, we found that transport is active, receptor-mediated and energy dependent We elucidated the characteristics of an oligonucleotide which are critical for uptake and studied the ability of certain oligonucleotide derivatives to compete for this uptake process. We found that methylphosphonates do not enter cells via this mechanism, but that phosphorothioates do, although much less efficiently than normal oligos. Uptake is by endocytosis and generally results in the occurrence of oligo-containing vesicles in the cytoplasm. In general, oligos are localized to cytoplasm and not the nucleus following uptake. By fluoresceinating novel oligo derivatives, one can easily follow their rate of accumulation, or lack thereof, by cells. In this way, more efficient oligos can be rapidly designed and tested. We have observed a novel intracellular oligonucleotide binding protein, different from the one we characterized on the cell surface, which may be involved in determining the intracellular fate of exogenously administered oligos. We are currently utilizing micropumps implanted subcutaneously in nude mice to explore the efficacy of constantly delivered antisense oligos as anti-tumor agents. To date, no toxicity due to oligo administration has been noted.